gganimate

Introduction

This tutorial goes through the fundamentals of the gganimate pacakge. This is an intermediate level workshop and assumes that learners are familiar with the “grammar of graphics” as implemented in ggplot2.

gganimate creates animated figures by transitioning plots through different “states.”

The key argument is the transition function.

We’ll start with a few R preliminaries

Preliminaries

set.seed(923741)

library("tidyverse")
library("gganimate")
library("gapminder")

transition_states()

transition_states() works with discrete data.

Goal: We want to create an animate figure which shows the Petal.Length and Petal.Width of the iris data set.

The first step to create an animated figure is to creat the static ggplot image. This will contain “all” the data to be plotted and will not look the same as the animated figure.

Load data

data(iris) # load the iris data
glimpse(iris) # view the data

## Observations: 150
## Variables: 5
## $ Sepal.Length <dbl> 5.1, 4.9, 4.7, 4.6, 5.0, 5.4, 4.6, 5.0, 4.4, 4.9, 5…
## $ Sepal.Width  <dbl> 3.5, 3.0, 3.2, 3.1, 3.6, 3.9, 3.4, 3.4, 2.9, 3.1, 3…
## $ Petal.Length <dbl> 1.4, 1.4, 1.3, 1.5, 1.4, 1.7, 1.4, 1.5, 1.4, 1.5, 1…
## $ Petal.Width  <dbl> 0.2, 0.2, 0.2, 0.2, 0.2, 0.4, 0.3, 0.2, 0.2, 0.1, 0…
## $ Species      <fct> setosa, setosa, setosa, setosa, setosa, setosa, set…

Static plot

Now create the static plot. Add any plot aesthetics at this point.

iris %>% 
  ggplot(aes(x = Petal.Width, y= Petal.Length, group = Species, color = Species)) +
  geom_point(size = 5, alpha = 0.75)

I like to build my plots step-by-step, so I save them as objects once I get them looking as I want.

iris_plot <- iris %>% 
  ggplot(aes(x = Petal.Width, y= Petal.Length, group = Species, color = Species)) +
  geom_point(size = 5, alpha = 0.75) +
  labs(x = "Petal width", y = "Petal length")

animated plot

With the static plot made, let’s add some animation with the transition_states() function. The user must declare the state to transition among.

iris_plot + 
  transition_states(Species)

Four arguments

transition_states() takes four arguments:

• states - variable name that contains discrete data
• transition_length - The relative length of the transition (default = 1)
• state_length - the relative length of the pause between states (default = 1)
• wrap - Should the animate wrap around? (default = TRUE) If TRUE the last state will be transitioned into the first.

transition_states(states, transition_length, state_length, wrap = TRUE)

These transitions look a bit slow. We have considerable power to fine-tune the speed of the transitions.

iris_plot +
  transition_states(Species,
                    transition_length = 0.1, 
                    state_length = 0.1)

labels

We can add labels!

• previous_state - previous variable state
• closest_state - the current variable state
• next_state - next variable state

If we are going to add the species name, I’ll go ahead and remove the legend since it would duplicate information in the plot.

iris_plot <- iris %>% 
  ggplot(aes(x = Petal.Width, y= Petal.Length, group = Species, color = Species)) +
  geom_point(size = 5, alpha = 0.75, show.legend = FALSE) +
  labs(x = "Petal width", y = "Petal length")

iris_plot +
  transition_states(Species,
                    transition_length = 0.1, 
                    state_length = 0.1) +
  labs(title = "Iris {closest_state}")

Easing

gganimate calculates intermediate (“tween”) values to transition among states. You can decide how this process occurs, which is called easing and can be adjusted with ease_aes().

ease_aes('linear'):

• quadratic: {^2} # add LaTeX here
• cubic: ^3
• quartic: ^4
• quintic: ^5
• sine: a sine function
• circular: 3
• exponential - ^e
• elastic - elastic release of energy
• back - pullback and release
• bounce - bouncing ball
  • methods
    • -in - function applied “as-is”
    • -out - function applied in reverse
    • -in-out - first half applied “as-is”, second half is reversed.

’cubic-in-out`

irisAnimation <- iris_plot +
  transition_states(Species,
                    transition_length = 2, 
                    state_length = 1) +
  labs(title = "Iris {closest_state}")

irisAnimation +
  ease_aes('cubic-in-out')

‘bounce-out’

irisAnimation +
  ease_aes(y = "bounce-out")

enter_ & exit_

A few of the fine tuning functions work for both transition_states() and transition_time().

Specifically we will deal with enter_ and exit_. Each can be paired with a variety of behaviors such as:

• appear / disappear - make elements appear and disappear at the start or end of their transition.
• fade - set alpha value to make elements fade in / fade out during transitions.
• grow / shrink - set elements to make them grow into or shrink out during their transition
• recolor - gradually change color or fill elements from their defined color to new color
• fly - set x and y positions where all elements enter from / exit to
• drift - modify the position of an element by a specified amount

Each has additional customizable parameters

transition_time()

`boxplots

Default is appear/disappear

anim <- ggplot(mtcars, aes(factor(gear), mpg)) + geom_boxplot() + transition_states(gear, 2, 1) # Fade-in, fly-out anim1 <- anim + enter_fade() + exit_fly(x_loc = 7, y_loc = 40) # Enter and exit accumulates anim2 <- anim + enter_fade() + enter_grow() + exit_fly(x_loc = 7, y_loc = 40) + exit_recolour(fill = ‘forestgreen’)